Brokaw (2003) The Poetics of Khipu Historiography

THE POETICS OF KHIPU HISTORIOGRAPHY:
Felipe Guaman Poma de Ayala's Nueva coronica

and the Relacion de los quipucamayos
Galen Brokaw*
State University of New York at Buffalo
By these knots they counted the successions of the times and when each Inca
ruled, the children he had, if he was good or bad, valiant or cowardly, with
whom he was married, what lands he conquered, the buildings he constructed,
the service and riches he received, how many years he lived, where he died,
what he was fond of; in sum, everything that books teach and show us was got
from there.^
Martin de Muriia
Abstract: The Andean khipu was a medium of colored knotted cords used to
record different types ofinformation in the pre-Columbian and colonial periods.
Although currently there is no way to read the khipu that have survived, numerous texts written in the colonial period claim to have relied on khipu as sources of
information. A comparison between two khipu transcriptions of Inca biographies on the one hand and the European biographical genre on the other reveal a
distinctly Andean poeticsin the sense ofa structural formatwith very suggestive links to semiotic conventions ofthe khipu.
INTRODUCTION
The Andean khipu is one of the most perplexing elements of indigenous Andean society. This device of colored, knotted strings has intrigued European observers from the first stages of contact through the
present. Although currently approximately 600 khipu are known to have
*The research for this essay was funded in part by a grant from the David Rockefeller
Center for Latin American Studies at Harvard University. I would like to thank Gordon
Brotherston, Kathleen Myers, Gary Urton, Jongsoo Lee, David E. Johnson, and LARR's
anonymous reviewers for their comments on earlier drafts of this article.
1. This and all other translations that appear in this article are my own. In many cases, I
have sacrificed eloquence and grammatical consistency in an attempt to reflect the nature
of the Spanish discourse to the degree possible. For the original Spanish text of this passage, see note 27.
Latin Atnerican Research Review, Vol. 38, No. 3, October 2003
2003 by the University of Texas Press, P.O. Box 7819, Austin, TX 78713-7819
112
Latin American Research Review
survived, the knowledge of how to read them has been lost, and exactly
what kind of information they hold and the nature of their semiotic function have been subjects of considerable debate. One of the questions
that colonial chroniclers attempted to answer about the khipu was
whether or not it constituted writing, and much of the debate today
centers around the same issue. Based on a selective and literal interpretation of colorual sources and a limited understanding of archaeological
specimens, many scholars have argued that the khipu was not writing
but rather a mnemonic device similar to a rosary. In the early twentieth
century, Leland Locke combined for the first time a systematic survey of
colonial records with a detailed analysis of actual khipu (Locke 1923).
Locke reinforced the same conclusions presumably reached by many
colonial chroruclers: that the khipu was merely numeric in nature, and
may have functioned as a mnemonic aid for more complex oral accoimts.
However, our understanding of the khipu as a semiotic medium is far
from complete (Urton 1994, 294), and sufficient ambiguity, contradiction, and general ignorance exist in the colonial chronicles to reject any
simple interpretation of statements made in such documents. More recently, Marcia Ascher and Robert Ascher have catalogued almost 200
khipu with detailed descriptions of each based on a common format and
analytic studies of their semiotic potential (Ascher and Ascher 1978,1981).
Contrary to what Locke maintained, Ascher and Ascher argue that the
khipu is perfectly capable of encoding detailed information, including
narratives.
One of the principal difficulties in assessing the function of the khipu
is not only a lack of iriformation but also the disparity between European and Andean cultural modes of thought and representation (Arellano
1999; M. Ascher 1986; Jara 1970; Prada Ramirez 1995,11; Quispe Agnoli
2002; Urton 2002). The cognitive negotiation involved in the dialogic
interaction between the acquisition of any secondary form of representation (i.e., alphabetic or khipu literacy) and the development of institutions that support it influences the way a culture conceives of the
signifying functions that it employs. Part of the problem is that the writing-orality opposition that tacitly underlies most discussions of the khipu
is problematic in this case (Arellano 1999; Quispe Agnoli 2002).^ Furthermore, the definition of writing itself is seldom questioned. Before
such issues can be resolved, or perhaps dismissed as irrelevant or unproductive, the nature of khipu semiosis must be understood.
In addition to the study of the archaeological khipu themselves, one
of the most productive approaches to understanding the khipu has been
the analysis of documents that contain information originally recorded
on khipu (Murra 1968,1981; Parssinen 1992,31-50; Rostworowski 1990;
2. See also Boone (1994), Brotherston (1992), and Mignolo (1994,1995).
KHIPU HISTORIOGRAPHY
II3
Rowe 1985; Urton 1998). In most cases, these writings are tribute records
that were recorded for official inspections {visitas) or for litigation in
court cases, but there are also narrative accounts that claim to have relied on khipu sources. Such documents are what we might call, modifying Mary Louise Pratt's term and its definition, textual contact zones:
textual spaces in which disparate modes of discourse meet, clash, and
grapple with each other in highly asymmetrical relations of domination
and subordination.^ All forms of discourse, whether oral or written, are
produced and received in the context of a genre or tradition (Bakhtin
1986; Foucault 1972), and they are tied in many ways to the medium
used to convey them. In this case, there are two discursive forces and
contexts that interact in the shaping of colonial documents derived from
khipu sources: the Andean and the Spanish.
The content of these documents was originally articulated in an
Andean context and through an Andean discourse linked in a dialogical
relatior\ship with Andean media (oral Quechua or Aymara and the khipu).
In the Andean context, a distinction between afc/ifpw-deriveddiscourse
and formal oral genres may not be justifiable. I would argue that the
dialogic relationship between the structure of Andean discourses,
Andean oral traditions, and the khipu was very different from the relationship between writing and orality in modem alphabetic cultures. It
may be that the khipu was not used to record all forms of oral discourse,
but all khipu were liriked to one or another oral genre. Also, as with alphabetic writing, secondary systems never replace, nor are they opposed
to, orality; rather they engage, supplement, and perhaps transform it
(Goody 1987,2000; McLuhan 1962; Olson and Torrance 1996; Ong 1982).
Two tjês of possible relationships exist between khipu and alphabetic texts from the colonial period that rely on khipu as sources of information. In the first case, close transcriptions of khipu, mostly tribute
records, were composed for official reports. This transcription process
involved the "reading" of the khipu in Quechua or Aymara by a
khipukamayuq, the transcription of this oral discourse into alphabetic
writing, and/or the translation of the indigenous language into alphabetic Sparush (Urton 1998,412). In the second case, the khipu is not transcribed, but serves as a source of information either directly or indirectly.
This is not to say that features of khipu textuality do not appear in such
texts, but that their influence may be much more subtle. Europeans who
claim to have relied on the khipu for information constructed histories
that consciously participate in European discursive genres, and, perhaps unconsciously, obliterate or deculturate many of the features that
3. In Imperial Eyes, Mary Louise Pratt defines contact zones as "social spaces where
disparate cultures meet, clash, and grapple with each other, often in highly asymmetrical relations of domination and subordination" (Pratt 1992, 4).
114
characterized the original indigenous text. Even though these writers

were not involved in the actual reading of khipu, the cohesiveness of
indigenous discourses would have resisted the complete assimilation to
European norms.
The extent to which khipu conventions survive the process of
transpositioning depends upon the degree of similarity between European and Andean genres, the cultural identity of the writers, their degree of education in or familiarity with the textual forms involved,
circumstances of production, ideology, and so on. As a product of an
educated, native Spaniard with adequate quantities of paper and time
to compose his history according to European standards, the nature and
degree of khipu textuality in Martin de Murua's Historia general del PerU
(1611), for example, is much more limited than in more direct khipu transcriptions of tribute records and historical works allegedly transcribed
directly from khipu like the Discurso sobre la descendencia y gobierno de los
Incas (1542/1608) or produced directly by indigenous writers who claim

to have relied on khipu sources like Felipe Guaman Poma de Ayala's
Nueva coronica y buen gobierno (c. 1615). If the writers or transcribers of
these texts used khipu as a primary source of information as they claim,
traces of the poeticsdefined in the original sense as the structural principles of a discourseôf the original khipu account should manifest
themselves in observable ways.
Any discussion of the relationship between a khipu that no longer
exists and its alphabetic translation/transcription is fraught with difficulties that stem from the lack of the original khipu text and our ignorance of the discursive conventions employed in different khipu genres.
Unfortunately, the Spaniards were not as interested in the khipu as a
material medium as they were in the Mesoamerican pictographic texts;
khipu were not considered exotic, collectible items like the codices; nor
were they preserved or studied in any great detail. With perhaps one
exception (Urton 2001), no one has ever matched up a specific khipu to a
specific alphabetic text. This makes it impossible to compare the meaning conveyed by an actual khipu text with that of an alphabetic one as
can be done with a pictographic codex.
Our ignorance about how the khipu worked and its capabilities as a
communicative medium,^ has, perhaps inevitably, caused discursive
analyses to neglect the intertextual relationship between the khipu and
4. For a definition and discussion of poetics and its meanings, see Todorov (1977,
1981), or browse the issues of the journal Poetics Today.
5. There is no way to even know for sure if we have examples of all the different
kinds/genres of khipu that existed in the pre-Hispanic and colonial periods. Indeed,
some have argued that because almost all extant khipu come from graves, they may
represent a very limited sampling of generic types.
II5
the written texts that relied on them for information.* At this point, the
best that we can do is to attempt to identify and analyze alphabetic texts
that are as close to direct transcriptions as possible. Studies of khipu transcriptions have been productive in several wâys. In "Las etnocategorias
de un khipu estatal," for example, through an analysis of a khipu transcription of tribute records, John Murra was able to identify a set of ethnocategories that reveal a hierarchical structure according to which Andean
society organizes the useful objects in the world around it (Murra 1981).
This discovery may be directly related to khipu conventions identified
as seriation (Altieri 1939; Radicati 1965), parallelism (Radicati 1965,1979),
format (M. Ascher 2002; Ascher and Ascher 1981, 81-107), patterns
(Conklin 1982, 271; Mackey 1990, 154; Urton 1994), relations (Ascher
and Ascher 1969, 1971), and structure (M. Ascher 2002). Furthermore,
this hierarchical structure demonstrates that, in this case at least, the
transcription process left intact a high degree of the structural integrity
of the original khipu account. In the investigation of a khipu-linked historiographic genre, the existence of multiple alphabetic texts containing
the same categories of information taken from different khipu sources
would provide a basis for a comparison and contrast between these texts
and similar European genres. I would argue that in this wây, it may be
possible to identify a trace of an Andean poetics that characterized the
original texts before they were transcribed, translated, and
transpositioned into alphabetic Spanish and assimilatedin the process
of reception and/or productioninto the closest equivalent European
discursive paradigm. Two of the most promising texts in this regard are
Guaman Poma's Nueva coronica and the work known as the Discurso de
la descendencia y gobierno de los Incas or the Relacion de los quipucamayos.''
6. Soon after the conquest of New World territories such as what is now Mexico and
Peru, Spanish officials began investigating indigenous customs, beliefs, and history. From
the very beginning, the Mexican codices were recognized as important documents that
recorded religious and historical information. The reaction to these documents, however, was ambivalent; most of them were burned, but the pictographic mode itself was
incorporated into official documentation. The indigenous society of Peru, governed by
the Incas at the time of the conquest, also had its own system of representation using
khipu, but, unlike Mexican pictography, this medium was never accepted into the official record. In "Representation in the Sixteenth Century and the Colonial Image of the
Inca," Tom Cummins explains that European memory "is conditioned by mimetic images in terms of both loci and imagines, and this is what could be read into the Mexican
pictorial manuscripts," but "the same could not be inscribed into the quipu.... Quipus
had no resonance with European forms, and while they were not categorized as creations of the devil, neither could they be adapted to simulate anything directly in the
Spanish colonial system. Thus, the quipu was transcribed into a European form simply
as an illustration or as a written text" (Cummins 1994,194-95).
7. This text is also known as Relacion de la descendencia, gobierno y conquista de los Incas.
ii6
TEXTS AND CONTEXTS
The Nueva coronica was written by Felipe Guaman Poma de Ayala,*

an indigenous Andean whose father belonged to a non-Inca noble family from Huanuco that had served as mitmaqkuna, groups relocated by
the Inca for royal service in newly conquered areas. He was bom sometime after the conquest in Huamanga, present-day Ayacucho, and he
died some time after 1615. Guaman Poma's work was inspired and
motivated by a combination of his experience working in the colonial
administration, his work for the chronicler Martin de Murua, and a
lengthy court battle over the rights to his family's land (Adomo 1993,
2001, 31-40). Although Guaman Poma relied extensively on European
writings for some portions of his text (Adomo 1986; Gabos-Fontana 2000;
Plas 1996), he also makes it very clear that his most important sources of
information were "the quipus and memories and reports of ancient Indians" [los quipos y memorias y rrelaciones de los yndios antiguos] (Guaman
Poma [1615] 1987,8[8]).

Pierre Duviols suggests that claiming a khipu source was a way of
garnering authority and attesting to the veracity of the text, and that as
a common stylistic device, it was an innocent fraud (Duviols 1979,589).
I would argue, however, that the convention of identifying khipu sources
is not as prominent as one would expect if it were seen as a guarantee of
authenticity or truth. In any case, the fact that it became a common device from a reader's point of view does not mean that it did not reflect
an actual relationship between khipu sources and alphabetic texts. The
fact is that the khipu was an ubiquitous device employed throughout the
Inca empire both before and after the conquest; the khipukamayuqs {khipumakers) were the people most likely to be interviewed by Spanish authorities and the most likely to be relied upon by indigenous writers.
Guaman Poma's illustrations demonstrate that at the very least he had
an intimate knowledge of how the khipu was handled and stored (Conklin
2002,53). In the Nueva coronica Guaman Poma does not identify the specific information derived from khipu sources, but many chrorucles reveal the different types of information stored on khipu. As a point of
departure, the following analysis takes Guaman Poma at his word, and
hypothesizes that much of the iriformation about indigenous Andean
history that appears in the Nueva coronica was collected either directly
or indirectly from khipu.
The second Andean text that will figure in the analysis below, the
Discurso sobre la descendencia y gobierno de los Incas (1542/1608), is more
8.1 am fully aware of the controversy caused by the Naples documents, but for now I
remain convinced that Guaman Poma was the author of the Nueva coronica. For a review
of the controversy, see Albo (1998) and Cantu (2001).
II7
explicit in terms of its relationship to khipu sources, but its provenience

is at the same time more problematic. The work, which was signed in
1608 by a Fray Antonio, can be divided into two sections (Porras
Barrenechea [1952] 1986; Duviols 1979; Urton 1990,44). The first section
consists mainly of a history of the Inca kings allegedly based on transcriptions of khipu records during an inquest conducted by the viceroy
Vaca de Castro in 1542 (Urteaga 1921, vi). The second section was probably written around 1608 and contains an accotmt of the events leading
up to and including the conquest. Raul Porras Barrenechea ([1952] 1986)
and Duviols (1979) have raised questions about the accuracy and authenticity of this document. Porras Barrenechea argues that the Fray
Antonio who signed the manuscript is Fray Antonio Martinez, who lent
support to legal petitions submitted by Don Melchor Carlos Inca (Porras
Barrenechea [1952] 1986, 749). The cormection between these individuals is significant because, as Duviols demonstrates, the purpose of the
Discurso was to establish two "facts," which correspond to the two sections of the text: (1) Don Melchor Carlos Inca's noble status as the legitimate heir of the Inca kings; and (2) the service rendered to the crown by
Paullu Topa Inca, the grandfather of Don Melchor Carlos Inca (Duviols
1979). The text, then, is essentially a probanza de meritos (proof of merit)
motivated by the possibility of receiving status and/or recompense from
the king (ibid, 587).
The portion of the text relevant for the analysis below is the first section containing the history of the Incas. According to the Discurso, the
viceroy Cristobal Vaca de Castro gathered together the oldest inhabitants of Cuzco and the surroimding area and asked them to give an accoimt of their origins and history. The responses that he received varied
greatly and often contradicted each other. Perceiving the frustration of
the Spaniards, these ancianos (elders) suggested that they seek out the
old khipukamayuqs. They explained that prior to the arrival of the Spaniards, Atahualpa had attempted to revise Inca history by burning all the
khipu he could find and killing the khipukamayuqs. Those that escaped
had been hiding in the mountains ever since. The viceroy sought out
these khipukamayuqs and had them brought to Cuzco with their khipu.
He then posed the questions to them, and had their accovmts translated
and transcribed.'
9. Al tiempo quegoberno en este reino del Peru el Licenciado Vaca de Castro, pretendiendo con
mucha solicitud saber la antigualla de los indios deste reino y el origen dellos, de los ingas,
senores quefueron destos reinos, y sifueron naturales desia tierra o advenedizos de otras partes,
para adveriguacion desta demanda, hizojuntar y parecer ante si a todos los ingas viejos e antiguos
del Cusco y de toda su comarca, e informandose dellos, como se prentendio, ninguno informo con
satisfaccion sino muy variablemente cada uno en derecho de su parte, sin saber dar otra razon
mas que todos los ingas fueron descendientes de Mango Capac, quefue el primero inga, sin saber
dar otra razon mas que todos los ingas fueron descendientes de Mango Capac, quefue el primer
ii8
Theoretically, the motivation for this historical inquiry was to determine whether or not the Incas were natural lords, which would in tum
validate or invalidate the right of the Spaniards to conquer and rule
them. But Duviols argues that the colonial ideology of the Spaniards
inevitably influenced the content of the document and the outcome of
the inquiry (Duviols 1979, 589). The ideology and personal interests of
the Discurso's author would have also influenced the production of his
own version. Although the original context of production as portrayed
in the document may be accurate, the version we have today was filtered through several processes: (1) transcription into alphabetic script
of an oral "reading" of a khipu record; (2) translation from Quechua into
Spanish; (3) modification by the ideology of Spanish conquestin this
case that of the 1540s or perhaps the 1580s; and (4) further modification
influenced by a combination of changing colonial and indigenous ideologies and the personal interests of the authorhere, probably Don
Melchor Carlos Inca or someone allied with him in the early 1600s.
I will discuss the ideology that informed the Discurso in more detail
below, but these processes impinge upon the production of practically
all texts produced by Spaniards or native Andeans who rely on
khipukamayuq informants. Duviols's concern is with the truth value of
the information contained in the Discurso or the degree of correspondence between the iriformation that appears in this document and that
of the original account. This is always an issue with any historical text,
but the essential question for the following analysis is not the "truth" of
the information or even whether it corresponds to the original khipu account but rather the degree to which any or all of the processes described
above caused a dissipation of the structure or format that characterized
the khipu-hased discourse of the original record. I am not so much interested in the information itself as I am in the categories of information
and their configuration or format within the discourse. Other than the
inga, sin saber dar otra razon, no conformando los unos con los otros. E vistose apurados en esta
demanda, dixieron que todos los ingas pasados tuvieron sus "quipocamayos", anst del origen y
principio dellos, como de los tiempos y cosas acontecidas en tiempo de cada senor ddlos; e dieron
razon que con la venida del Challcochima e Quisquis, capitanes tiranos por Ataovallpa Inca que
destruyeron la tierra, los cuales mataron todos los quipcamayos que pudieron haber a las manos y
les quemaron los "quipos", diciendo que de nuevo hablan de comenzar (nuevo mundo) de Ticcicdpac
Inga, que ansi le llamaban a Ataovallpa Inga, e dieron noticia (de) algunos que quedaron, los
cuales andaban por los montes atemorizados por los tiranos pasados. Vaca de Castro envid por
ellos, y le trujeron antel cuatro muy viejos.
Estos quipocamayos habian sido a manera de historiadores o contadores de la razon, y fueron
muchos, y en todos ellos habia conformidad en sus quipos y cuentas; no tenian otro ejercicio mis
de tener gran cuenta con sus quipos ansi del origen y principio de los ingas, como de cada uno en
particular, desde el dia que nascian cada uno, como de las demas cosas acontecidas en tiempo de
cada senor dellos. (CoUapifia and Supno [1542/1608] 1921,3-5)
II9
inherent distortions caused by the translation and transcription themselves, there is no indication that there would have been any specific
ideological reason for modifying the basic structure of the Inca biographies that appear in the Discurso.
I am not merely taking for granted that the structure of the Discurso
reflects in some way a corresponding khipu structure that may have survived in some form. Rather, this is the hypothesis that I will attempt to
demonstrate. The starting point is the generally accepted assertion
even by Duviols and Porras Barrenecheathat the first part of this text
was originally transcribed from khipu (Duviols 1979, 589; Porras
Barrenechea [1952] 1986, 748; Urton 1990,45). Absent the original khipu
there is no way of examining the specific transcultural processes involved
in the reading, transcription, translation, and editing of the 1542 inquest
that informed the first section of the Discurso. But a comparison and
contrast with the European tradition and other Andean texts such as
Guaman Poma's Nueva coronica makes it evident that at least a residue
of a distinctly Andean poetics survives the translation.
One of the main projects in the textual analysis of indigenous Andean
chronicles such as Guaman Poma's Nueva coronica and, although to a
much lesser extent, the Discurso, has been the identification and analysis of their European sources or influences (Adomo 1986; Cabos-Fontana
2000; Lopez Grigera 2001; Perez Canto 1996; Plas 1996; Cummins 1994,
196). In many cases, the discussion of European sources is limited to
textual models, because the content itself is of Andean origin. Often, the
implicit assumption in the identification of European models for indigenous writings is that the authors of these texts took an indigenous content and expressed it using a European form. None would argue that
there was no Andean form that structured the information prior to its
transcription and translation into alphabetic Spanish, but the significance
of this fact is routinely ignored. Andean reality, like that of the rest of the
Americas, was certainly a world upon which the Spaniards imposed
many forms; the cognitive-discursive paradigms that the Europeans had
developed to perceive, describe, discuss, and argue about the world predetermined the parameters within which the raw non-discursive material of the Americas would emerge (O'Gorman 1958). Yet European
writers could not observe Inca history first-hand. They had to rely on
native informants who employed indigenous discourses and had no
other choice but to engage the paradigms presented to them through
these discourses.
This is not to say that European discursive structures had no influence, but that European subjects were forced to negotiateprobably
subconsciously in most casesthe terms of any representation derived
from native sources. Every history of the Inca empire written during the
colonial period has a distinctly biographical component that to a greater
120
or lesser extent structures the accoimt (Julien 2000). The identification of

these biographical texts with the European tradition claims to explicate
the productive influence of the Spanish biographical genre (Adomo 1986,
40-47; Cummins 1994, 196). However, this is as much, or more, a "receptive transformation" in which the texts are perceived through a preestablished literary-cognitive paradigm as it is a recogrution of any real
adherence to familiar European genres.
This "receptive transformation" relates to what James Lockhart calls
"double mistaken identity" in which relationships of similarity based
on common denominators of sociocultural phenomena from two different cultures lead each to equate a foreign practice with its own; and
what Martha James Hardmari-de-Bautista identifies as the translation
tradition:
[a] mutual, if unspoken, agreement from both sides, such that certain items in
one language (both words and grammatical forms) were translated by specific
items in the other, such that these agreed upon translations came to be believed
to be the "true" and "only" correct translations. (Hardman-de-Bautista 1982,153)
Here, however, we are not dealing with cultural concepts, discrete grammatical structures, or even individual words but rather extended discursive forms. The result is often not a direct substitution of one form
for another (except perhaps in the reception of these texts), but rather an
ad hoc, hybrid product with unique characteristics that were never assimilated or generalized over time into a distinct and stable written genre.
European and highly acculturated writers such as Martin de Murua and
the Inca Garcilaso de la Vega'" assimilated and transformed Andean discourses to a greater degree than indigenous writers such as Guaman
Poma or the translators of the history related by the khipukamayuqs
CoUapifia, Supno, and others in the Discurso. Before analyzing these texts
in detail, however, a critical understanding of the European biographical genre is necessary as a basis for comparison in order to reveal and
avoid the pitfalls of the translation tradition.
HISTORY AND BIOGRAPHY IN THE EUROPEAN TRADITION
Prior to the conquest of the Americas, Spanish historiographers were

not producing general histories of the type exemplified by Alfonso el
Sabio's General historia (13* c.) or the Estoria de Espana (13* c.) (Sanchez
Alonso 1941). This may have been due to the popularity and availability
of several versions of such histories in both Latin and Spanish, and to
the reliance on these established textual authorities. Furthermore, in
10. For an analysis of the convergence of indigenous and European voices in Garcilaso's
text, see Mazzotti (1996).
121
"Historiografia medieval: constantes evolutivas de un genero," Fernando

Gomez Redondo explains that in the second half of the fourteenth century at the end of the reign of Alfonso XI, two changes occurred in the
thought or mentality of Spanish society: (1) a growing disinterest for the
remote past of the nation as it had been reflected in the Estoria de Espafia
and its posterior derivatives; and (2) the life of the monarch acquires a
great importance as the protagonist of history and as a sign of his time
(Gomez Redondo 1989,4). Gomez Redondo explains that these changes
led to the conversion of the cronica general into the cronica real. Renaissance humanism inherits and adapts this emphasis on the individual,
incorporating it into its historiographical methodology. Historians began focusing on major figures of Spanish society, presenting descriptions and biographies as exempla to be emulated or shurmed.
During the age of discovery and conquest, however, writers were
forced to deal with the fact that the older histories made no mention of
the Americas. This embarrassing gap in the supposedly universal historical record induced a revival or rejuvenation of the historia general in
the sixteenth and seventeenth centuries. Yet the old models had been
weakened, and the general flexibility and mixing of literary genres in
the Renaissance affects historiography as well (Sanchez Alonso 1941,
355-463; Colie 1973). The humanistic biographical genre in particular
provided a convenient structure for the seemingly fantastic exploits of
conquistadors with often dominant personalities. The two most important works that set the standard for biographical histories and which
might have influenced New World historiography are Feman Perez de
Guzman's Generaciones y semblanzas (1450) and Hemando del Pulgar's
Glaros varones de Castilla (1486) (Tate 1965, xxii-iii). Educated writers
would have been familiar with the tradition of biographical histories,
and indigenous and mestizo writers who were not as versed in European textual traditions may have become familiar with this format either indirectly through the chronicles written in the Andes, or directly
through such works as Diego Rodriguez de Almela's Valerio de las estorias
escoldsticas e de Espafia (1487), which seems to have been a popular work
imported in the latter half of the sixteenth century (Adomo 1986, 43;
Leonard 1942,23)."
In "Sobre la biografia espanola del siglo XV y los ideales de la vida,"
Jose Luis Romero explains that the Spanish biography exhibits three main
structural features: (1) the lineage of the individual; (2) a physical and
moral description; and (3) significant accomplishments (Romero 1945,
52-61). In the introduction to his edition of Generaciones y semblanzas.
11. Although the Valerio is biographical, it is structured thematically around such topics as justice, miracles, and various virtues and vices (Rodriguez de Almela [1487] 1994,
3-5) in which biographical anecdotes provide exempla for merely illustrative purposes.
122
Robert Brian Tate observes a slightly different, but still tripartite, set of
elements that structure Perez de Guzman's biographies: (1) lineage of
the subject; (2) a description of physical characteristics, temperament,
and moral qualities; and finally (3) the date and place of death as well as
the age of the deceased (Tate 1965, xvii). Tate explains that this formula
is most obvious in the shorter biographies like that of Don Alfonso
Enriquez:
Don Alfonso Enrriquez, admiral of Castile, was the bastard son of don
Fadrique, Master of Santiago, son of King Alfonso.
He was a man of medium height, white, red, thick in the body, not too
bright, but discrete and prudent, very graceful in speech. He was quick to anger
and often caught up in it; of great strength, friend of the honorable, and those
who were of royal lineage but of lesser estate found favor and help from him.
He had an honorable house, he set a good table, he understood more than he
spoke.
He died in Guadalupe at the age of seventy-five years. (Perez de Guzman
[1450] 1965,14-15)"
This biography does not contain the anecdotes/exemplfl that are so prevalent in the Valerio, but other biographies in the collection do include more
extensive narratives. This bare-bones version reveals the humanist emphasis on the individual as the protagonist of history. In this case, the
description of the subject itself serves as an example. The content of histories dedicated to kings has a more political character, but even so the
focus remains on the ruler as an individual who is converted into a sign
of the times (Gomez Redondo 1989,4).
Table 1 represents a consolidation of the elements identified by Romero
and Tate. This table outlines the discursive paradigm of the E!uropean
biographical genre. Theories of genre, language, discourse, and cognition have exposed an inherent relationship between the elements of discursive structures and cognitive processes (Foucault 1972; Niles 1999,
12-13; Turner 1996)." Writers who set out to compose a text can never
be completely original: to one degree or another, they must imitate other
12. Don Alfonso Enrriquez, almirante de Castilla, fue fijo bastardo de don Fadrique, maestre
de Santiago, fijo del rey don Alfonso.
Fue onbre de mediana altura, bianco, roxo, espeso en el cuerppo, la razon breve e corta, pero
discreta en atentada, osoz grof ioso en su dizir. Turbavase muy a menudo con safia e era muy
arrebatado con ella; degrande esfuergo, de buen acogimiento a tos buenos, e los que eran linaje del
rey e non tenian tanto estado, fallavan en el favor e ayuda. Tenia honrrada casa, ponia muy buena
mesa, entendia mas que dizia.
Murio en Guadalupe en edat de setenta e ginco anos. (Perez de Guzman [1450] 1965,14-15)
13. We could even go back as far as Ren6 Descartes and Francis Bacon. In Novum
Organum (1620), for example, Francis Bacon identifies four impediments to human understanding: idols of the tribe, of the cave, of the market-place, and of the theater (Bacon
[1620] 1994, 53-68). These idols consist of preconceptions, habits, and dogmas invariably related to, and perpetuated by, language.
I23
Table 1 Discursive Paradigm ofthe European Biographical Genre

1. Lineage
2. Physical and moral description
3. Significant accomplishments
4. Death
texts that they have read (Frye 1964, 40). And, in any case, in the sixteenth and seventeenth centuries, originality was not understood as it is
today. Adherence to the principle of imitatio was a conscious aspiration
and the criterion for excellence." Thus, although the elements are general categories of information, the paradigm was a fairly rigid model
that we may use to evaluate the relationship of Andean texts to this
tradition.
HISTORY AND BIOGRAPHY IN THE NUEVA
COR6NICA
Undeniably European discursive genres influenced Guaman Poma's

Nueva coronica y buen gobierno. The overall form of the work is cast in the
genre of an historia general. As explained above, the incorporation of a
biographical mode within an historia is not remarkable, but it stands out
here because the text does not synthesize the genres. Inca Garcilaso de
la Vega's Comentarios reales (1609) also combines biography and the more
general history,^^ but this text, written about the same time and covering
much of the same material as the Nueva coronica, weaves the individual
life histories of the Inca emperors into a much broader account of Andean
history and culture, a form more in tvme with the historiographic conventions of the time. Evidently Guaman Poma's European education
was not what it might have been. His limited exposure to the textual
tradition of Europerelative to Garcilasomay have contributed to the
"disorgaruzation" and "unpolished" nature of the work. Many of these
apparently unorthodox features, however, are consistent with Andean
structures of meaning. The organization based on five world ages, for
example, or the division of Andean society into ten age-graded census
groups of men and ten of women reflect the structure of indigenous
discourses (Brokaw 2002; Prada Raniirez 1995,23; Rowe 1958; Ossio 1970,
1973; Wachtel 1973). At first glance Guaman Poma's Inca biographies
also appear to derive from European models (Adomo 1986,4047), but
closer analysis suggests that there is more at work here.
14. For a discussion of the principle of imitatio in Renaissance Humanism, see Streuver
(1970,144-154).
15. Garcilaso, however, begins with a brief discussion of the geographical status of the
New World and then immediately focuses on Peru. He omits the traditional beginning
based on the Biblical account in "Genesis."
124
Any given genre or discursive formation is tied to an ideology. The

emergence of a biographical historiography in Europe was linked on
the one hand to a redistribution of emphasis between divine and secular authority (Gomez Redondo 1989; Sanchez Alonso 1941), and on the
other to the humanist interest in the individual. The content and structure of these biographies were also influenced by other ideological systems such as aristocratic lineage and moral virtues. These ideologies
and the force of the textual tradition irifluenced by them are precisely
what has led critics to read the Inca biographies that appear in colonial
chronicles as manifestations of the European biographical genre. However, the Andean ideology of dynastic descent associated with what is
known as capac status more adequately accounts for the specific nature
of these texts.
In Reading Inca History (2000), Catherine Julien explains that capac was
a hereditary status with accompanying privileges and rights granted to
the descendants of the first Inca, Manco Capac, through the male line
Qulien 2000,23-48). The Inca ruler himself possessed the highest degree
of capac status, which legitimized his authority. But other individuals
and groups also held different degrees of capac status that determined
their rarO< in the Inca hierarchy. In this situation, political struggles often involved competing claims about complex genealogical relationships.
Julien explains that formalizing an account of dynastic descent was a
necessity in order to avoid false claims and confusion in the calculation
of capac status. For this reason the memory of Manco Capac's descendants was strong "even after the argument that the Incas were powerful
because they were capac no longer made any sense." (Julien 2000,17)
In other words, the ideological weight of genealogically determined
capac status as a source of political legitimacy permeated all levels of the
Inca empire both before and after the conquest. The importance of capac
status, then, was one of the major ideological determinants of Andean
historiographic genres; the possibility of fraudulent genealogical claims
would have been a major motivation for fixing these accounts in khipu.
Julien "excavates" numerous documents in an attempt to get at the
underlying Andean genres that informed Spanish accounts of Inca history. She argues that colonial texts reveal two distinct but related genres
that she labels the "genealogy" and the "life history." The genealogy
was a series of short accounts that might include brief narratives, but
was concerned primarily with dynastic origin and the establishment of
the panaca of each Inca generation (Julien 2000, 89). The life history is
also genealogical in nature, but it included much more extensive information, especially for later rulers (91). In spite of their differences, the
ideology of capac status that informs both genres clearly differs from the
ideology of European nobility evident in Spanish biographies.
The first Inca biography in Guaman Poma's Nueva coronica contains
I25
introductory material not present in the other biographies. Perhaps not

the best example for a comparison of features characteristic of the genre,
it nevertheless reveals an important difference in genealogical conceptions. The following passages from the European text Generaciones and
the Nueva coronica illustrate the differences:
Generaciones:
This King, Don Enrique the Third, was son of King Johan and Queen Leonor,
daughter of King Pedro of Aragon, and descendant of the noble and very ancient and illustrious generation of Gothic Kings and especially of the glorious
Catholic prince Ricaredo, King of the Goths of Spain. And according to the histories of Castille, the blood of the Kings of Castille and their succession from
one King to the next has continued until today, which are more than eight hundred years without interruption by any other line. Which I believe will be found
in few lines of Christian kings that last so long a time. [...]
And this King Enrique began to reign at a little over eleven years of age,
and reigned sixteen, thus he lived more than twenty-seven years. (Perez de
Guzman [1450] 1965,45)'*
Nueva cordnica:
From the first Inca Manco Capac, who reigned one hundred and sixty years
with the beginning and with the last Topa Cuci Gualpa Huascar legitimate Inca,
and of his bastard brother Atahualpa Inca and between the time the said Incas
first started to reign and the end when their kingdom was ended and consumed,
the said legitimate by law ruled one thousand five hundred and fifteen years by
governing the land the said Incas and kings. (Guaman Poma [1615] 1987,
87[87])"
The biography of Enrique III begins by establishing his lineage following the formula identified by Tate and Romero. The nobility of Enrique
Ill's line is derived from the illustrious reputation of his ancestors whose
origins are located in the vague reference to the "generation of Gothic
Kings." The longevity of the line is more a source of prestige than a
legitimizing agent. To a large extent, legitimacy was taken for granted;
it was honor and prestige that concerned Spanish biographers. At first
16. Este rey don Enrique el tergero fue fijo del rey don Johan y de la reyna dona Leonor, fija del
rey don Pedro de Aragdn, e desgendio de la noble e muy antigua e clara generation de los reyes
godos e sefialadamente del glorioso y catolico principe Recaredo, rey de los godos en Espana. E
segunt por las estorias de Castilla paresge, la sangre de los reyes de Castilla e su sugesion de un
rey en otro se ha continuadofasta oy, que son mas de ochogientos anos sin aver en ella mudamiento
de otra Una nin generacidn. Lo qual creo que sefallard en pocas generagiones de los reyes christianos
que tan luengo tienpo durase. [...]
E este rey don Enrique comengo a reinar de poco mds de onze afios, e reino diez e seis, assi que
bivid mds de veinte e siete aiios. (P^rez de Guzman [1450] 1965,4-5)

17. Desde el primer Ynga Manco Capac que rreyno ciento sesenta afios con el comienso y con
el postrer Topa Cuci Gualpa Uascar Ynga lexitimo y de su ermano uastardo Atagualpa Ynga y
desde que eomenso a rreynar los dichos Yngas y acabar su rreyno, como se acabo y consumio su
rreyno, los dichos lexitimos de derecho que rreynaron mil y quinientos y quinze anos de senorear
en la tierra estos dichos Yngas y rreys. (Guaman Poma [1615] 1987, 87[87])
126
glance the passage from the Nueva cordnica may seem similar to the biography in Generaciones in that it deals with the lineage of rulers descended from Manco Capac. However, the ideology that informs the
continuity of Inca rule in this passage is neither honor nor prestige but
rather legitimacy as established by genealogically determined capac status. Furthermore, although this passage from the Nueva cordnica appears
on the same page, it does not form part of Manco Capac's biography.
This short section functions as an introduction to all of the biographies
and relates to the empire rather than to the biographical structure. The
biography proper of Manco Capac immediately follows this introductory segment, and it conforms to the same model as the rest of the Incas.
Julien asserts that Guaman Poma's biographies do not appear "to
have reproduced Inca genres to any degree" (59), but she seems to base
this argument on his genealogical iiiformation, which often differs from
that of other chroniclers. The issue here, however, should not be whether
or not the information is corroborated by other chrorucles, but rather
whether the iriformation categories correspond to those of similar texts.
Approached from this perspective, Guaman Poma's biographies should
be classified under what Julien identifies as the genre of dynastic genealogy. I would argue that what Julien calls the genealogical gerure might
be better characterized as genealogical biography. The history is genealogical in the sense that the individual biographies do not appear in
isolation. The genealogical link is important because it legitimizes each
successive Inca by establishing that he possesses the highest degree of
capac status, but it is only one of many essential elements that comprise
this biographical structure. A thorough analysis of these individual biographies will help further illuminate their relationship to European
and Andean traditions.
Although some variation is evident, the biographies of the Incas in
the Nueva cordnica exhibit a highly formulaic quality (Mroz 1984,1989).
Ten basic elements constitute Guaman Poma's biographical format: (1)
a description of the ruler's clothing and arms; (2) physical appearance;
(3) life works, achievements, conquests, or the way he influenced the
empire by instituting traditions, conquering territories, and so on; (4)
the name of the Inca's coya (queen); (5) the ruler's life-span and death;
(6) legitimate children; (7) illegitimate children; (8) legacy or something
memorable about the Inca; (9) the cumulative number of years ruled by
the Incas up to that point; and (10) the successor to the Inca. All of these
biographies appear on a single page following a drawing of the ruler.
The shortest biography, dedicated to the fourth Inca Mayta Capac, illustrates the formula. I have labeled each of the different elements with
sequential numbers in brackets:
[1] His weapons and helmet uma chuco were dark blue yanas pacra and his masca
paycha and conga cuchuna, ualcanca and his cloak blood-red and his singlet blue
I27
on top with three rows of tocapu and on the bottom boxes with white, green, and
red, and four sandal cords. [2] And he was a very ugly man in his face, feet,
hands, and body, thin, shivery, very afflicted. Nevertheless, valorous, melancholic. [3] In addition to his father's kingdom, he conquered up to Potosi and
Charcas and many provinces and towns. [4] And he was married to Chinho
Urma Mama Yachi. [5] And he died in Cuzco at the age of one hundred and
twenty years [6] and he left riches to his idol Guana Cauri. [7] And he had children Chinho UcUo Mama Caua, Apo Maytac Inca, Vilcac Inca, Uiza Topa Inca,
Capac Yupanqui Inca, Curi UcUo. [8] And he had other bastard sons auquiconas
and bastard daughters nustaconas who were very many in number. And he had
a daughter that he loved very much, and he called her Inquillay Coya. [9] Four
Incas ruled five hundred sixty-five years. [10] Afterwards, his legitimate son
Capac Yupanqui Inca succeeded him. (Guaman Poma [1615] 1987,99[99])'^
Table 2 lists the essential components of Guaman Poma's biographies

and the order in which they appear. The first element in Guaman Poma's
biographical format provides a description of the Inca's arms and clothing. I will discuss the significance of this item in more detail below. Here,
it is important to note that no precedent in the European tradition exists
for the inclusion of this type of information.
The second element in Guaman Poma's biographical structure is a
description of the Inca's physical appearance and in a few cases, his character.^' This seems to coincide with the Spanish model (table 1, no. 2), but
unlike the Sparush version, Guaman Poma's text does not link the physical qualities of the subject with his personality in a distinctly moralizing
tone. The Nueva cordnica describes Uira Cocha Inca, for example, as:
"Handsome man, white of body and face, and he had a sparse beard,^
and he had a good heart" [Gentil hombre, bianco de cuerpo y rrostro y tenia
18. [1] Tenia sus armas y zelada uma chuco de azul escuro yanas pacra y su masca paycha y
conga cuchuna, ualcanca y su manta de encarnado y de su camexeta de hazia arriua azul y del
medio tres betas de tocapo y de auajo caxane con bianco y uerde y Colorado y quatro ataderos de
los pies. [2] Y fue muy feo hombre de cara y pies y manos y cuerpo, delgadito, friolento, muy
apretado. Con todo eso, brabicimo, melancoUco. [31 Y conquisto demds que tenia su padre hasta
Potociy Charca y muchas prouincias y pueblos. [4] Yfue cazado con Chinho Urma Mama Yachi.
[5] Y murio en el Cuzco de edad de ciento y ueynte anos [6] y dejo rriquiesas a su ydolo Guana
Cauri. [7] Y tubo hijos ynfantes Chinho UcUo Mama Caua, Apo Maytae Ynga, Vilcac Ynga,
Uiza Topa Ynga, Capac Yupanqui Ynga, Curi Ucllo. [8] Y tubo otros hijos bastardos auquiconas
y hijas bastardas nustaconas quefueron muy muchos. Y tenia una hija que le queria muy mucho,
y aci le llamo Ynquillay Coya. [9] Reynaron quatro Yngas quinientos y sesenta y cinco anos.
[10] Despues susedio su hijo lexitimo Capac Yupanqui Ynga. (Guaman Poma [1615] 1987,
99[99])
19. All of the Inca biographies Include a brief physical description except for that of
Manco Capac, the first Inca.
20. The expression pocas barbas could refer to someone with little knowledge and experience {Diccionario de autoridades), but the context of this passage seems to suggest a
literal meaning. And the accompanying drawing of this Inca in Guaman Poma's text
depicts him with stubble on his chin (Guaman Poma |1615] 1987,106[106]).
128
Table 2 Narrative Outline of Guaman Poma's Biographies

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Distinctive dress of the Inca

Physical appearance and character
Life works; achievements; conquests
Coya (queen)
Life span; death
Legacy or memory
Legitimate children
Illegitimate children
Cumulative time of Inca rule
Successor to the Inca
Unas pocas barbas y tenia buen corazdn] (Guaman Poma [1615] 1987,107[107]).
No real discussion of temperament takes place, and the moralizing tone

is absent as well. The text does often include damning or redeeming qualities such as Inca Roca's confiscation of the property of the poor, or Yauar
Uacac Inca's favorable disposition toward the poor, but it does so without the explicit moral overtones. This and the brevity of the accounts limit
the exemplary value that the biographies might have had. Although these
biographies certainly portray the Irica as a sign of his time, they do so in
a way much different from the European model.
The remaining categories of information in the discursive formation
derived from Guaman Poma's biographies might seem to fall within
the expectations of the European biographical paradigm, but only if the
second element, "significant accomplishments," is understood in the very
broadest of terms. The inclusion of "illegitimate" children, for example,
is not characteristic of the European genre; nor is the subject's spouse a
standard component. The Andean model is a highly structured sequence
of very specific informational categories. A comparison of tables 1 and 2
reveals significant differences.
The contrast, then, between the discourse of European biographies
and that employed in the Nueva cordnica reveals the divergent nature of
these texts: they are informed by different ideologies; they convey different types of information; the components that they have in common
often function differently; and the moralizing tone of the Spanish gerure
is absent in the Andean version. Although this does not, in and of itself,
establish a khipu origin for the text, the high degree of consistency in
format throughout the biographies is not consistent with the generally
more variable nature of strictly oral genres. The rigidity of the model
implies the possible influence of some kind of material medium that
functions in a way similar to writing. The implication here is not only
that the Nueva cordnica biographies represent a uruque Andean gerure,
but also that this ger\re was determined in many ways by the semiotic
conventions of the khipu. Although Guaman Poma is not specific about
I29
the information he took from khipu sources, his biographies are very
similar to those found in the Discurso, which clearly identifies its own
origin in khipu.
HISTORY AND BIOGRAPHY IN THE DISCURSO
Like Guaman Poma's record, the Discurso presents the history of the
Inca empire by sketching the biographies of each individual ruler These
biographies exhibit a highly formulaic structure that is consistent
throughout. The sections dedicated to Lloque Yupanqui and Mayta Capac
illustrate the biographical format of the Discurso:
CLnchiroca was succeeded by his son Lluque Yupangue Inga. He did not
increase the kingdom because during his reign there were many rebellions by
those whom he had inherited, and he was at the point of losing the kingdom; he
did everything he could to maintain that which he had inherited from his fathers. And his woman was Mama Caba. With her he had three sons: the oldest
was Mayta Capac; the second was Apo Conde Mayta; the third Apo Taca. From
these younger sons descended those from the Chigua Yuin ayllu. He ruled more
than fifty years.
Lluqui Yupangui Inga was succeeded by Mayta Capac Inga, who did not
increase the kingdom in any way, because he was always at war with his own
subjects, who were rising up in rebellion every day. And his woman was Mama
Taoca Ray; and with her he had two male children: the older and his successor
was called Capac Yupangui Inga, the younger Apo Tarco Guaman. From this
younger son descended those of the ayllu Uscamaitas. He reigned fifty years.
(CoUapifia and Supno [1542/1608] 1921,13)^'
These two biographies follow a format that characterizes the presentation of all twelve Inca rulers: (1) the Inca is presented as the successor to
his father; (2) military conquests and the area that he governed; (3) the
Inca's spouse; (4) the oldest male child; (5) the younger children and the
ayllu of their descendants; and (6) the number of years during which the
Inca reigned. The history of each Inca is not limited to this iriformation,
but these elements are the essential common denominators that appear
in every section and, with very few exceptions, in the same order. Table 3
represents the structure of the Discurso's biographical format.
21. A Cinchiroca subcedio su hijo Lluque Yupangue Inga. Este no aumento porque en su
tiempo tuvo muchas rebuliones de los que habia heredado, e tuvo el SeHorio en puntos de perder;
harto hizo en sustentar lo que de sus padres habia heredado. E tuvo por mujer a Mama Caba.
Tuvo en ella tres hijos: el mayorazgofue Mayta Cdpac Inga; el segundo fue Apo Conde Mayta; el
tercero Apo Taca. Destos menores descienden los del ayllo Chigua Yuin. Reino mds tiempo de
cincuenta afios.
A Lluqui Yupangui Inga subcedid Mayta Capac Inga, el cual no aumento cosa alguna, porque
siempre tuvo guerra con los suyos, que cada dia se le alzaban. E tuvo por mujer a Mama Taoca
Ray; e tuvo en ella dos hijos varones: el mayor y subcesor se llamd Cdpac Yupangui Inga, el
menor Apo Tarco Guaman. Deste menor descienden los del ayllo Uscamaitas. Este reino cincuenta
afios. (CoUapina and Supno [1542/1608] 1921,13)
130
Table 3 Format of the Discurso

1.
2.
3.
4.
5.
6.
Names ruler as successor to previous emperor

Conquests, achievements
Woman {coya, queen)
Oldest male child
Younger children and the ayllu of their descendants
Length of rule
Although a European reading of the Discurso may identify it with the

Spanish biographical tradition, the same incongruities with that tradition exhibited by Guaman Poma's work appear here as well. As in the
Nueva cordnica, for example, the moralizing tone characteristic of European texts such as Generaciones y semblanzas is conspicuously absent from
the Discurso, and the same structural disparities are evident. However,
like the Nueva cordnica, the biographical model employed in the Discurso
exhibits a high degree of stability that suggests an established textual
tradition. It is not unreasonable to conclude, therefore, that this work
derives these features from the conventions of the khipu, especially since
the text identifies itself as a khipu transcription.
This conclusion gains support not only from the Discurso's lack of
compliance with the European model and its formal stability, but also
from its close correlation to Guaman Poma's format. Table 4 provides a
comparison and contrast between the biographical structures derived
from European biographies and the format derived from the two Andean
texts discussed here. A few similarities between the purely European
paradigm and the format of the Nueva cordnica and the Discurso are evident, but they are fairly weak. The differences far outweigh them. The
similarities between the two Andean texts, on the other hand, suggest a
generic relationship. The Nueva cordnica and the Discurso differ, but a
degree of variance is to be expected in any given generic tradition. The
processes of transpositioning and translation that mediate these works
would orJy exacerbate the differences.
The Nueva cordnica, for example, ends each biography by naming the
successor to the Inca, who will then appear in the following biography.
The Discurso, on the other hand, places this element at the beginning of
each biography. Although the Discurso may seem to parallel the Spanish
format, which begins by establishing the lineage of the subject, the function of these elements clearly differs from one to the other. In the Spanish tradition, lineage functions in many respects as a measure of
individual worth. In the Discurso, on the other hand, just as in the Nueva
cordnica, the statement of succession is a device that establishes political
continuity, confirms capac status, and functions as a discursive transi-
I3I
Table 4
European biography
Nueva coronica
1. Lineage
2. Physical and
moral description
3. Significant
accomplishments
4. Death
Discurso
1. Successor of previous Inca
1. Dress of the Inca

2. Physical appearance
3. Life works/conquests
4. Coya [queen]
5. Life span; death
6. Legacy or memory
7. Legitimate children
8. Illegitimate children
9. Cumulative Inca rule
10. Successor to the Inca
2. Conquests/achievements
3. Woman [coya, queen]
4. Oldest male child
5. Younger children & ayllu
6. Length of rule
[see #1]
tion between the end of one biography and the begirming of another
often with no reference to any family relationship, which is already established in element number 7 of the previous Inca's biography.
Furthermore, the position of this transitional element at the beginning of the biography in the Discurso may be misleading. In most cases,
the phrase that constitutes this component is a short, independent statement: "Cinchiroca was succeeded by his son Lluque Yupangue Inga" [A
Cinchiroca subcedid su hijo Lluque Yupangue Inga] (13); "Mayta Capac was
succeeded by his son Capac Yupangui" [^4 Mayta Cdpac Inga subcedid
Cdpac Yupangui su hijo] (13); "Capac Yupangue was succeeded by Inga
Roca" [A Cdpac Yupangue subcedid Inga Roca] (14). Some of the biogra-
phies link the transitional phrase to the following biography: "Lluqui

Yupangui Inga was succeeded by Mayta Capac Inga, who did not increase the kingdom in any way" [A Lluqui Yupangui Inga subcedio Mayta
Cdpac Inga, el cual no aumentd cosa alguna] (13). This conjimction though
might have been the work of the transcriber/translator. Horacio

Urteaga's modem edition of the Discurso places these statements at the
beginning of the biographies, and indicates the division through the use
of paragraphs, but in the original discourse the statements of succession
may have belonged conceptually to the preceding biography. This would
be consistent with Guaman Poma's format in which the successor is
named after reporting the cumulative years of Inca rule at the end of
each biography: "Four Incas ruled five hundred sixty-five years. Afterwards, his legitimate son Capac Yupanqui Inca succeeded him" [Reynaron
quatro Yngas quinientos y sesenta y cinco anos; despues susedid su hijo lexitimo
Capac Yupanqui Ynga] (92); "Five Incas ruled seven himdred and five
years and Ynga Roca began his reign" [Reynd cinco Yngas setecientos y
132
cinco anos y sucidid Ynga Roca] (94). Further corroboration for this argument is found in the captions to the Gilcrease Inca portraits, which also
follow this pattern, naming the Inca's successor at the end of each text
(Bames 1998,238-42).
The biggest difference between the two paradigms is the difference
in the number of information categories. The Discurso biography seems
to be a somewhat abbreviated version of the Andean biographical genre
that appears in the Nueva cordnica. Several possible explanations may
account for this disparity. The khipu source of the Discurso may have
contained more categories of ii\formation that were left out of the reading, the transcription, or the translation. The official inquiries that motivated the transcription of this information were normally based on a set
of questions sent from Spain. The khipukamayuqs, CoUapina, Supno, and
the two others, may have performed an abbreviated reading in response
to the specific information requested by the Spanish. It is equally as plausible that the Spaniards who were transcribing and translating the khipu
readings (Vega 1974,14-15) would have edited out any information that
deviated too far from what they felt was relevant.
A few items in each of the models, although very similar, differ sUghtly.
Items seven and eight of the Nueva cordnica paradigm differ from the corresponding items four and five of the Discurso, but there is a kind of equivalence in that both make a distinction between two groups of siblings. The
ninth position in the Nueva cordnica refers to the cumulative years of Inca
rule, while the corresponding sixth element in the Discurso only mentions
the length of the reign of an individual Inca. Here again, this may be the
result of Spanish editing or regional variations of khipu conventions. The
khipukamayuqs who produced the Discurso were from Pacariqtambo
(Cuzco), and Guaman Poma was from Lucanas (Ayacucho).
The differences may also stem from different reading practices or preferences. The information stored on a khipu is not displayed in a linear
fashion. The khipukamayuq was presented with various dimensions of
information organized according to a set of conventions that allowed
numerous reading options. The difference between the Nueva cordnica's
cumulative years of Inca rule and the Discurso's reign of an individual
Inca ruler, for example, may simply reflect the result of a calculation
made by Guaman Poma or a desire to emphasize one detail instead of
another. I would argue that the items of infonnation provided in the
two texts are similar enough to consider them general manifestations of
the same generic categories.
One possible explanation for the similarity between these two works
is that Guaman Poma was familiar with the text of the Discurso and
used it as a source in writing the Nueva cordnica. If this were the case,
one would expect not only the format but also the specific information
I33
in the biographies to be the same or very similar Similarity here is relative, because most of the major figures of Inca history would inevitably
be the same in all accounts. To complicate matters, even if Guaman Poma
did rely upon a version of the Discurso, he might have modified some of
the information to suit his own political interests. Although the structure of informational categories is almost identical in these two texts,
significant differences in the information itself arise. However, these differences have no apparent ideological significance for either Melchor
Carlos Inca or Guaman Poma. A comparison between the versions of
Mayta Capac's biography in each text illustrates the difference.
Nueva cordnica:
[. . .]In addition to his father's kingdom, he conquered up to Potosi and Charcas and
many provinces and towns. And he was married to Chinho Urma Mama Yachi. And
he died in Cuzco at the age of one hundred and twenty years and he left riches
to his idol Guana Cauri. And he had children Chinho IJcllo Mama Caua, Apo
Maytac Inca, Vilcac Inca, Uiza Topa Inca, Capac Yupanqui Inca, Curi Ucllo. And
he had other bastard sons auquiconas and bastard daughters nustaconas who
were very many in number. And he had a daughter that he loved very much,
and he called her Inquillay Coya. Four Incas ruled five hundred sixty-five years.
Afterwards, his legitimate son Capac Yupanqui Inca succeeded him. (Guaman
Poma [1615] 1987,99[99]; my emphasis)^
Discurso:
Lluqui Yupangui Inga was succeeded by Mayta Capac Inga, who did not increase
the kingdom in any way, because he was always at war with his own subjects,
who were rising up in rebellion every day. And his woman was Mama Taoca Ray;
and with her he had two male children: the older and his successor was called
Capac Yupangui Inga, the younger Apo Tarco Guaman. From this yoimger son
descended those of the ayllu Uscamaitas. He reigned fifty years. (CoUapifia and
Supno [1542/1608] 1921,13; my emphasis)
Thus, neither the Nueva cordnica nor the Discurso conforms to the European biographical genre, but they both independently exhibit the same,
highly stable discursive format. This evidence lends support to their
own claims of khipu origins. The structural stability both within and
between these texts suggests a link to some form of secondary material
medium as opposed to a strictly oral transmission. If we accept that the
provenience of the biographies in the Discurso can ultimately be traced
22. y conquisto demds que tenia su padre hasta Potociy Charca y muchas prouincias y pueblos. Yfue cazado con Chinho Urma mama Yachi. Y murio en Cuzco de edad de ciento y ueynte
afios y dejo rriquiesas a su Guana Cauri, Y tubo hijos ynfantes Chinho Ucllo Mama Caua, Apo
Maytac Ynga, Vilcac Inca, Uiza Topa Ynga, Capac Yupanqui Ynga, Curi Ucllo. Y tubo otros
hijos bastardos auquiconas y hijas bastardas nustaconas quefueron muy muchos. Y tenia una
hija que le queria muy mucho, y aci le llamo Ynquillay Coya. Reynaron quatro Yngas quinientos
y sesenta y cinco afios. Despues susedid su hijo lexitimo Capaca Yupanqui Ynga. (Guaman
Poma [1615] 1987,99[99])
134
back to one or more khipu, then based on structural similarities the same
is probably true of the Nueva cordnica biographies. All of the above supports the assertions made in the texts themselves, and suggests that we
should take such claims seriously.
In Julien's comprehensive survey of colonial chronicles, she argues

that the genre of genealogical biography, of which the Nueva cordnica
and the Discurso are examples,^ does "not appear to have relied to any
great degree on quipo recording" Qulien 2000,49). Rather, she maintains
that there is a strong possibility that these accounts were recorded
through paintings rendered on wooden tablets (56-59, 89). A pre-Hispanic pictographic tradition of Inca portraiture seems to have existed
(Gisbert 1980,117), which may or may not have been different from pictographic narratives described by chroniclers like Cristobal de Molina
and Bemabe Cobo (Molina [1576] 1988,49-50; Cobo [1653] 1956,59). Yet
limiting a particular Andean genre to one medium may be misleading.
The confusion arises in part because different chronicles that seem to
incorporate the same Andean genre often explicitly identify different
media sources, pictographic in one case and khipu in another One solution to this problem is to sin\ply disbelieve or attempt to discredit the
provenience of one of the texts. I would argue that a more likely explanation is that the genre could be encoded in more than one medium. In
the case of the Nueva cordnica and the Discurso, however, the primary
medium seems to have been khipu.
Although each of Guaman Poma's biographies is accompanied by a
fuU-page drawing of the Inca to which it corresponds, it is very clear that
these illustrations contain no narrative information. The only formal link
between the illustrations and the corresponding text is the physical description that appears as the second element in the format outlined above.
It seems likely that if Guaman Poma had relied on a pictographic narrative as the source for this portion of his work, he would have reproduced,
at least in part, the narrative nature of the original drawings. These Inca
drawings in the Nueva cordnica may reflect a more iconic genre of Inca
portraiture that Guaman Poma incorporated verbally into his primarily
Wj/pu-derived account. Eliminating in this way the physical description of
the Incas from the strictly khipu-dehved text brings the biographical format of the Nueva cordnica even closer to that of the Discurso.
23. Julien's analysis is somewhat ambivalent about the Nueva cordnica. She discusses it
along with other genealogies, but with regard to both Martin de Murua and Guaman
Poma she states that "[n]either author appears to have reproduced Inca genres to any
degree, although a memory of the painted tradition may have inspired their work" (Julien
2000, 59).
I35
Even so, the khipu could have recorded visual information. Fernando
Prada Ramirez suggests that the distinctive features of dress that characterized the various ethruc groups in the Inca empire were used to identify them in the khipu records kept by royal administrators (Prada Ramirez
1995, 28). The chronicler Pedro de Cieza de Leon explains that these
administrators
went from town to town looking at the dress of the inhabitants and the resources
they had and the size of the land or if they had livestock
Those who are sent
to assess the provinces, entering into one, where they see by the quipos the inhabitants that there are. (Cieza [1553] 1985,74)^^
Antonio Vazquez Espinoza states that these differences in dress existed prior to the Inca conquest and that they were allowed to persist in
order to facilitate ethnic identification (Vazquez Espinoza [1630] 1992,
743), but Cieza de Leon seems to suggest that the Inca imposed a dress
code on the populations that they conquered (Cieza [1553] 1985, 163,
166). The Inca may have imposed specific features of ethnic dress in
order to facilitate khipu record keeping, and it would have been fairly
easy to encode such information on a lihipu.
The absence of such descriptions in other khipu-hased chronicles such
as the Discurso may be due to selective readings or translations. Given
that the European paradigm did not include a description of clothing,
many chroniclers would not have felt compelled to include such information in their accounts. It would have been natural for them to filter
out what they saw as irrelevant. Indigenous chroruclers who might have
understood the significance of these descriptions may have left them
out because they knew they were irrelevant to the Spaniards, or the khipu
upon which Guaman Poma relied might have been more elaborate variations of the genre.
In contrast to the Nueva cordnica, the Discurso contains no physical
descriptions or other indicators that it was linked to a pictographic text.
On the contrary, it explicitly identifies its provenience in khipu. Julien
disregards this claim in the Discurso in part because the origin myth of
the Inca that precedes the series of biographies includes elements that
only emerged in late sixteenth-century documents, which suggests that
the text was not transmitted pristinely from the original khipu record
(Julien 2000, 64). The scope of my argument does not include origin
myths; rather it focuses exclusively on the Inca biographies themselves.
In the Discurso, the origin myth does not form part of the biographical
structure. Manco Capac is the main protagonist of the origin myth, but
24. Iban de pueblo en pueblo mirando el traje de los naturales y posibilidad que tenian y la
grosedad de la tierra o si en ellas habia ganados . . . . Visitando los que por los Incas son enviados
las provincias, entrando en una, en donde ven por los quipos la gente que hay (Cieza [1553]
1985, 74).
136
his biography, as defined by the format outlined in table 4, appears immediately following the origin story, followed by the biographies of the
subsequent Incas. Furthermore, the fact that the original account was
contaminated in some way does not negate the significance of the khipu
as the original source in terms of the possible persistence of structural
features tied to khipu conventions.
One of the most suggestive structural features of Guaman Poma's
biographical format is that the number of components constitutes a complete decimal unit of ten. The decimal unit was-and still isan important ordering principle in Andean society (Urton 1997), informing social
organization and structures of meaning (Brokaw 2002; Julien 1988;
Wachtel 1971, 127-30). I have argued elsewhere that this principle of
decimal organization was intimately tied up with the conventions of
khipu semiosis (Brokaw 2002). It would be tempting to argue that each
of these categories corresponds to a string or a group of strings on a
khipu. But, here, we must not prematurely attribute this organization to
a manifestation of an authentic indigenous conceptual system based on
a decimal structure. These categories are a provîsional, ad hoc, heuristic
device that I have devised to facilitate my analysis. In "Paracronologia
dinastica de los Incas segiin Guaman Poma," Marcin Mroz divides these
same biographies into nine elements,^^ and other scholars will undoubtedly break them up in still different ways.^* Furthermore, in the case of
the Discurso biographies, it is impossible to arrive at a decimal structure.
The point is that no matter how they are divided, a consistent, stable
pattern emerges in which a specific number of components appear in
each biography in the same order with a low degree of variance.
Although the ten categories I have identified in Guaman Poma's biographies may not reflect the Andean principle of decimal organization, other features do. In both the Nueva cordnica and the Discurso, the
biographies that vary the most from the format outlined in table 4 are
the final two dedicated to the eleventh and twelfth Incas, Hayna Capac
and Huascar Gordon Brotherston argues that the first ten Incas represent "a decimal subset... whose reigns (unlike those of the last two ...)
exceed normal human life spans and esoterically link the world ages
with modern history" (Brotherston 1992,81[81]). This difference may be
due to a standardization of khipu historiography precisely during the
reign of the tenth Inca, Tupac Yupanqui. Although the khipu had existed
25. Mr6z divides the biographies in the following way: (1) royal attributes, arms, and
dress; (2) physical aspect and character; (3) relationship to gods, idols, huacas; (4) conquests; (5) corresponding coya (queen); (6) death, place, and age; (7) legitimate sons and
daughters; (8) illegitimate children; and (9) aggregate number of years transpired (Mroz
1989,25).
26. In my own initial analysis of this text, I identified seven categories of information
in Guaman Poma's biographies (Brokaw 2001).
I37
for hundreds of years before the Incas, the Nueva cordnica explains that
Tupac Yupanqui instituted the widespread use of the khipu in the Inca
adn\inistration:
And he began to organize his estate, as well as communities and storehouses
throughout the kingdom using accounts and khipu.... And he had assessors
yncap rantin rimac; attorneys and protectors runa yanapac; secretaries yncap
quipocnin; scribes Tahuantinsuyo quipoc; accountants hucha quipoc. (Guaman
P o m a [1615]
1 9 8 7 , [ ] 2 '
This passage suggests that it was probably under Tupac Yupanqui's leadership that the khipu was adapted for historiographic use by the secretaries {yncap quipocnin) and scribes {Tauantinsuyo quipoc). This would
explain why the history of the Incas includes ten rulers precisely at that
point. The importance of the decimal unit would have made a decimally
structured historical paradigm highly desirable.^* If Tupac Yupanqui were
responsible for the institution of khipu historiography, the more consistent structure of the first ten biographies would make sense: these accounts would have been composed at the same time on khipu hy the
same khipukamayuq or group of khipukamayuqs, while the biographies of
the subsequent Incas would have been added later, perhaps by different
khipukamayuqs. In the interim, the khipukamayuqs may have developed
more advanced techniques of khipu semiosis, and the generic model itself might have changed slightly over time in response to the demands
of the Inca. Furthermore, the last two Incas were the most recent, and
first-hand experience with the physical personage of the Inca as well as
the historical events were still present in the memories of the living inhabitants of Tawantinsuyu. The more recent accounts would be more
prone to supplementation and/or elaboration based on the narrator's
own experience or common knowledge.
Other variations or omissions that occur in the genealogical biographies may be due to the processes of transcription and translation of the
original khipu texts, but they may also be the natural result of variables
in the construction and reading of the khipu themselves. One of the least
stable elements in Guaman Poma's biographies is the sixth category that
I have labeled "legacy or memory." This item may come before the
legitimate children (no. 1, Manco Capac; no. 2, Cinchi Roca; no. 8, Huira
Cocha Inca; no. 9, Pachacuti Inca Yupanqui; and no. 10, Topa Inca
Yupanqui), between the legitimate and the illegitimate children (no. 6,
27. Y comensd hazer su hazienda y comunidad y depocitos con mucha horden, qilenta y quipo
en todo el rreyno. . . .Y tubo azesor yncap rantin rimac; procurador y protetor runa yanapac;
secretario yncap quipocnin; escribano Tauantinsuo quipoc; contador hucha quipoc (Guaman
Poma [1615] 1987, 111 [111]).

28. For an analysis of the use of decimal organization in Guaman Poma's historical
paradigm, see Brokaw (2002).
138
Inga Roca), after the illegitimate children (no. 4, Mayta Capac; and no. 5,
Capac Yupanqui Ynga), or not at all (no. 3, Lloqui Yupanqui Ynga; and
no. 7, Yauar Uacac Ynga). The absence of any given information category does not cause significant problems. The organization of cords
and colors on actual khipu employ conventions that allow for the unambiguous omission of any given element within a sequence while maintaining the integrity of the pattern (M. Ascher 1983, 274; Ascher 1986,
270; Radicati 1979,88-89).
The few variations in the position of biographical elements in the
Nueva cordnka as well as in the Discurso may relate to different modes of
khipu reading. The khipu consisted of a horizontal main cord to which
were tied vertical pendant cords. The pendant cords are divided into
groupings indicated by such conventions as spacer-cords, color differentiation, and spatial separation. Each of the pendant cords may have
its own configuration of subsidiary cords attached to it. These subsidiary cords, in turn, may have their own subsidiary cords, and so on.
Knots may be tied into the cords at all levels. Thus, in reading the khipu,
one might read horizontally or vertically (Mroz 1984, 87; Rowe 1985,
197). A horizontal reading moves from one pendant cord to the next,
then returns to the subsidiary cords at the end of a pendant cord grouping if more detail is desired. A vertical reading would articulate the information on the pendant cord, then its subsidiaries before moving on
to the next pendant. The complex three-dimensional nature of khipu texts
made possible a variety of conventional reading styles or techniques for
which there is no parallel in the linear system of alphabetic reading.
These different reading techniques may even account for the differences
between what Julien identifies as genealogical and life-history genres:
an abbreviated, horizontal reading might correspond to tbe more precise, less elaborate genealogical biographies, and the vertical reading
would then produce the longer, more detailed life-histories.
Admittedly, this analysis only focuses on two texts, but the format
that structures these textsor traces of itis also evident in many other
chronicles from the colonial period. The structural features of this distinctly Andean genre are evident to one degree or another in all versions
of Inca history. Martin de Munia actually explains that the history of the
Inca empire recorded on khipu took a biographical form:
By these knots they counted the successions of the times and when each Inca
ruled, the children he had, if he was good or bad, valiant or cowardly, with
whom he was married, what lands he conquered, the buildings he constructed,
the service and riches he received, how many years he lived, where he died,
what he was fond of; in sum, everything that books teach and show us was got
from there. (Murua [1611] 1986,373)^
29. Por estos nudos contaban las sucesiones delos liempos y quando reind cada ynga, los hijos
I39
Mtmia is not consciously attempting to analyze the poetics of khipu historiography, but he provides a list that foUows very closely the structiure of
Guaman Poma's biographical paradigm presented in table 2.^ This is not
surprising given the close connection between the works of these two
chroniclers (Mendizabal Losack 1963; Ballesteros Gaibrois 1978,1981; Ossio
1998). The evidence suggests that MiHiia and Guaman Poma may have
relied on the same khipukamayucjs in the composition of their accounts.
Although Julien's comparison of niunerous Andean chronicles focuses
primarily on the relationship between texts based on the content of information categories, her discussion demonstrates that the categories
themselves are fairly stable (Julien 2000). And the analysis of the individual texts of the biographies from the Nueva coronica and Discurso confirms that the sequential order of these information categories is
consistent as well. This kind of stability is normally associated with discourses caught up in a dialogical relationship with some kind of secondary medium such as writing (Brokaw 2002, 287). This is precisely
what one would expect from a discourse based on khipu conventions of
seriation, cord configuration, color patterns, decimal organization, and
so on that have been identified both in actual khipu (Ascher and Ascher
1969,1971,1978,1981; Conklin 1982; Mackey 1990; Radicati 1965; Urton
1994,2001) and in khipu transcriptions orfc/ifpu-derivedtexts (Murra 1981;
Brokaw 2002).
The consistent ordering of information categories in the Andean genre
of the genealogical biography may have functioned similarly to the structure of ethnocategories that characterize the format of the khipu analyzed by Murra (1981): a stable hierarchy of elements that determine the
order in which information categories appear on khipu. Andean
ethnocategories used in tribute records may actually have been a technique developed dialogically between the cultural conceptualization of
the social and natural world on the one hand and the conventions of the
khipu on the other. In the case of khipu historiography, the order of the
biographical categories may have been more of a semiotic convention
than a widespread cultural mode of thought, but this is the type of conventional adaptation that one would expect in the development of khipu
que tubo, sifue bueno o malo, valiente o cobarde, con quienfue casado, que tierras conquistd, los
edificios que labro, el sirbicio y riqueza que tubo, quantos anos biuio, donde murio, a que fue
aficionado; todo en fin lo que los libros nos ensenan y muestran se sacaba de alii y ansi (Murua
[1611] 1986,373).
30. Guaman Poma also criticizes other chroniclers who "get it wrong." Speaking of
Domingo de Santo Tomas, Guaman Poma complains: "And he did not write the genealogy of the first Indians, how, and in what way the pacarimos [natives] lived and multiplied or about their lives" [Y no escriuid la desendencia de los primero yndios, como, de que
manerafiie y multiplied antiguamente de tos primeros senores, rreys, pacarimos [originario] y
de SUS uidas] (Guaman Poma [1615] 1987,1079[1089]).
140
semiosis as it expanded to accommodate the needs of the Inca social

and political machine. I would argue, therefore, that what we know about
the capacity of the khipu and its use in storing historical iriformation
points to this medium as the stabilizing force behind the Andean genre
of biographical history. What emerges from these records, then, is a distinct poetics of khipu historiography.
CONCLUSION
In the context of a cultural contact zone such as the colonial Andes,

we must be extremely careful in identifying cultural products with
particular traditions. I have argued here that the apparent conformance
of the biographical format evident in the Nueva coronica and the Discurso
to many features of the European biographical genre does not, in and of
itself, indicate that they derive from the same tradition. The nature of
the subject matter and the superficially universal aspects of many human life experiences such as birth, death, and sexual relationships make
inevitable certain coincidences that will be reflected in the discursive
modes developed in different cultures. In producing their texts, Guaman
Poma and the khipukamayuq informants of the Discurso conceived of their
projects according to the terms of the Andean culture that don\inated
their conceptions of the world and representations of it. The medium
that gives us access to this information, however, was foreign to, and
incompatible with, these indigenous Andean conceptions. Inevitably,
both readers and writers in such a situation will attempt to establish
points of contact between the two paradigms, an interface through which
a compromise can be reached between the semiotic demands of one culture and the medium of another. In this case, the interface is the sinularity between the biographical modes of Andean and European textual
traditions. We cannot say, therefore, that the European biography did
not influence the Nueva coronica, the Discurso, and similar Andean texts,
for it may very well have done so. However, both of the textual traditions involved also influence the reading of the text. For modem readers, the European discursive modes are the most evident because the
text is composed in Spanish, and without a cultural sensitivity to Andean
textual conventions only the genre of European biography is perceptible. An understanding of the Andean textual tradition provides the
necessary cognitive framework for at least the intellectual recognition
of the under- and overlying dimensions derived from khipu.
The question remains about the nature of khipu semiosis, which is not
the same as the debate about the khipu as writing. In anthropology, writing has been a benchmark used to measure the level or nature of human
civilizations. Cultures have often been classified as those with (historic)
and those without (pre-historic) writing. This scheme creates an opposition between writing and orality that has no room for alternative forms
of representation. Faced with an other mentality, an other "literacy" based
on a technology that establishes a different relationship between medium and discourse, the European colonial episteme can only understand this other in terms of an already known, in-between category:
mnemonics. Indeed, the most common analogue used to describe the
khipu is the mnemonic rosary. An understanding of the khipu requires a
deconstruction of this writing-(mnemonics)-orality opposition. Whether
or not the khipu constitutes writing is not the issue I have pursued here.
That is ultimately a semantic and, hence, political problem that may be
solved by expandingor notthe definition of writing. But there are
several dimensions of the khipu that converge in such a way as to suggest that this medium of knotted, colored string is much more than a
mnemonic device.
Undeniably the khipu employed a set of highly complex conventions
capable of encoding semasiographic or even phonographic iriformation
(cord configuration, numeric quantities, extra-numeric knots, colors and
color patterns, ethnocategories, etc.), and these features would only have
developed in response to a semiotic desire or need. Abundant testimony
from the colonial period claims that the khipu was a narrative encoding
device and transcriptions of khipu historiography such as the Discurso
and Guaman Poma's Nueva coronica attest to the existence of highly stable
genres of discourse. Furthermore, the structural features of this discourse
exhibit a close correlation to known semiotic conventions of khipu.
The analysis presented above contributes to a general understanding
of the relationship between khipu, the Andean genre of biographical history, and alphabetic texts that relied upon information derived from this
medium. It is important to differentiate this type of analysis, however,
from what Robert Ascher calls encipherment: a thought experiment that
hypothesizes about how specific information was encoded in khipu; a
kind of back formation that attempts to match up the structure of information with actual khipu conventions (R. Ascher 2002, 108-111).'' Although I have briefly mentioned general conventions that may be
relevant in this area, the primary project here has been a comparative
discourse analysis tbat identifies the nature of khipu-linked Andean
genres as they have survived in colonial writings. The analysis suggests
that the historiographic khipu stored information in a non-mnemonic
way that when "read" produced a generic series of Inca biographies.
31. For another example of this encipherment methodology, see Parssinen (1992, 3150).
142
Although this type of comparative analysis is very productive, it is

currently limited by the relatively small number of actual khipu transcriptions that have been discovered. Julien's work demonstrates that
the structure of Andean discursive genres permeate the chronicles written in the colonial period (Julien 2000). Unforttmately, very few texts
like the Discurso can be identified as originatingregardless of whatever discursive or ideological filters through which they may have
passedin a direct khipu transcription. There are surely many more relevant documents lying undiscovered in archives tbat will cast further
light on the nature of khipu semiosis, and perhaps provide a firmer basis
for Ascher's encipherment methodology. The first step is to recognize
that the material medium of the khipu was linked to genres of Andean
discourse, powerful discursive paradigms tbat do not simply dissolve
and disappear when translated and transpositioned into Spanish.
REFERENCES
ADORNO, ROLENA
1986
1993
2001
Guaman Poma: Writing and Resistance in Colonial Peru. Austin: University of Texas
Press.
"The Genesis of Felipe Guama Poma de Ayala's Nueva coronica y buen
gobiemo." Colonial Latin American Review 2 (1-2): 53-92.
Guaman Poma and His Illustrated ChroniclefiromColonial Peru: Prom a Century of
Scholarship to a New Era of Reading. Copenhagen: Museum Tusculanum Press.
ALB6, XAVIER
1998
La Nueva coronica y buen gobiemo: lObra de Guaman Poma 0 de jesuitas?

Anthropologica 16:307-48.
ALTIERl, ANDRfe RADAMES
1939
"Sobre un kipu peruano." Revista de Antropologia de la Universidad Nacional de

Tucuman 1 (1): 7-13.
ARELLANO HOFFMAN, CARMEN
1999
"Quipu y tocapu: Sistemas de comurdcacion Inca." In Los Incas: Arte y simbolos,

edited by Frariklin Pease et al., 215-62. Lima: Banco de Cr^dito del Peru.
ASCHER, MARCIA
1983
1986
2002
"The Logical-Numerical System of the Inca Quipus." Annals of the History of

Computing 5 (3): 268-78.
"Mathematical Ideas of the Incas." In Native American Mathematics, edited by
Michael P. Closs, 261-89. Austin: University of Texas Press.
"Reading Khipu: Labels, Structure, and Format." In Narrative Threads: Accounting and Recounting in Andean Khipu, edited by Jeffrey Quilter and Gary Urton,
87-102. Austin: University of Texas Press.
ASCHER, MARCIA, AND ROBERT ASCHER
1969
1971
1978
1981
"Code of Ancient Peruvian Knotted Cords (Quipus)." Nature 222: 529-33.

"Numbers and Relations from Ancient Andean Quipus." Archive for History of
Exact Sciences 8 (4): 288-320.
Code ofthe Quipu: Databook. Ann Arbor: University of Michigan Microfilms.
Code ofthe Quipu: A Study of Media, Mathematics, and Culture. Ann Arbor: University of Michigan Press.
ASCHER, ROBERT
2002
"Inka Writing." In Narrative Threads: Accounting and Recounting in Andean Khipu,

edited by Jeffrey Quilter and Gary Urton, 103-15. Austin: University of Texas
Press.
I43
BACON, FRANCIS
1994
Novum Organum. Translated and edited by Peter Urbach and John Gibson. Chicago: Open Court.
BAKHTIN, MIKHAIL
1986 [1620] Speech Genres and Other Late Essays. Austin: University of Texas Press.
BALLESTEROS GAIBROIS, MANUEL
1978-79 "Relaci6n entre Fray Martin de Murua y Felipe Huaman Poma de Ayala." In
Amerikanistische Studien: PestschriftfiirHermann Trimborn/Estudios americanistas:
Libro jubilar en homenaje a Hermann Trimborn, edited by Roswith Harman and
Udo Oberem, 1:39-47. St. Agustin, Germany: Anthropos Institut.
1981 "Dos cronistas paralelos: Huaman Poma y Martin de Murua (confrontacion de
las series reales graficas)." Anales de literatura hispanoamericana 9 (10): 15-66.
BARNES, MONICA
1998
"The Gilcrease Inca Portraits and the Historical Tradition of Guaman Poma de
Ayala." In Andean Oral Traditions: Discourse and Literature, edited by Margot
Beyersdorff and Sabin Dedenbach-Salazar Saenz, 223-56. Bonn: Bonn
Americanist Studies 24.
BOONE, ELIZABETH HILL
1994
"Introduction: Writing and Recording Knowledge." In Writing without Words:

Alternative Literacies in Mesoamerica arid the Andes, edited by Elizabeth Hill Boone
and Walter D. Mignolo, 3-26. Durham: Duke University Press.
BROKAW, GALEN
2001
2002
"El khipu como fuente en la Nueva coronica de Felipe Guaman Poma de Ayala."
In Guaman Poma y Bias Valera: Tradicion Andina e Historia Colonial, edited by
Francesca Cantu, 417-29. Rome: Instituto Italo-Latinoamericano.
"Khipu Numeracy and Alphabetic Literacy in the Andes: Felipe Guaman Poma
de Ayala's Nueva coronica y buen gobiemo." Colonial Latin American Review 11, no.
2:275-303.
BROTHERSTON, GORDON
1992
The Book ofthe Fourth World: Reading the Native Americas Through Their Literature.
Cambridge: Cambridge University Press.
CABOS-FONTANA, MARIE-CLAUDE
2000
Memoire et acculturation dans les Andes: Guaman Poma de Ayala et les influences
europeennes. Paris: L'Harmattan.
CANTO, FRANCESCA
2001
Guaman Poma y Bias Valera: Tradicion andina e historia colonial. Rome: Instituto
Italo-Latinoamericano.
CIEZA DE L E 6 N , PEDRO DE
1985 [1553] E; senorio de los Incas. Madrid: Historia 16.

COBO, BERNABfi
1956 [1653] Historia del Nuevo Mundo: Obras del Padre Bemabe Cobo, Biblioteca de Autores
Espafioles, vols. 91-92. Edited by Francisco Mateos. Madrid: Ediciones Atlas.
COLIE, ROSALIE
1973
"Inclusionism: Uncanonical Forms, Mixed Kinds, and Nova Reperta." In The Resources of Kind: Genre-Theory in the Renaissance, edited by Barbara K. Lewalski,
76-102. Berkeley: University of Califorrua Press.
COLLAPINA, AND SUPNO, ET AL.
1921 [1542/1608] "Discurso sobre la descendencia y gobierno de los Incas." In

Informaciones sobre el antiguo Peru, edited by Horacio H. Urteaga, 1-53. Lima:
Sanmarti y Cia,.
1974 [1542/1608] Relacion de la descendencia, gobierno y conquista de los Incas. Edited by
Juan Jos^ Vega. Lima: Ediciones de la Biblioteca Universitaria.
CONKLIN, WILLIAM
1982
2002
"The Information System of Middle Horizon Quipus." Ethnoastronomy and Archaeoastronomy in the American Tropics, edited by Anthony F. Aveni and Gary
Urton, 261-81. New York: New York Academy of Sciences.
"A Khipu Information String Theory." In Narrative Threads: Accounting and Recounting in Andean Khipu, edited by Jeffrey Quilter and Gary Urton, 53-89. Austin: University of Texas Press.
144
^^*'" American Research Review
CUMMINS, TOM
1994
"Representation in the Sixteenth Century and the Colonial Image of the Inca."
In Writing Without Words: Alternative Literacies in Mesoamerica and the Andes, edited by Elizabeth Hill Boone and Walter D. Mignolo, 188-219. Durham: Duke
University Press.
DUVIOLS, PIERRE
1979
"Datahon, paternity et ideologie de la 'Declaraci6n de los quipucamayos a Vaca

de Castro' (Discurso de la descendencia y gobiemo de los tigas)." In Les cultures iberiques en devenir: essais publies en hommage a la memoire de Marcel Bataillon
(1895-1977), 583-91. Paris: La Fondation Singer-Polignac.
FOUCAULT, MICHEL
1972
The Archaeology of Knowledge. New York: Pantheon Books.
FRYE, NORTHROP
1964
The Educated Imagination. Bloomington: Indiana University Press.
GARCILASO DE LA VEGA, EL INCA
1991 [1609] Comentarios reales de los Incas. Edited by Carlos Aranibar. Lima: Fondo de
Cultura Econ6mica.
GISBERT, TERESA
1980
lconografia y mitos indigenas en el arte. La Paz: Editorial Gibert y Cia.
G O M E Z R E D O N D O , FERNANDO
1989
"Historiografia medieval: constantes evolutivas de un g^nero." Anuario de

Estudios Medievales 19: 3-15.
GOODY, JACK
1987
2000
The Interface Between the Written and the Oral. Carhbridge: Cambridge University
Press.
The Power ofthe Written Tradition. Washington, D.C: Smithsonian Institution Press.
GUAMAN POMA DE AYALA, FELIPE
1936 [1615] Nueva coronica y buen gobiemo (Codex peruvien illustre). Travaux et Mtooires
de L'Irwtitut d'Etnologie, vol. 23. Paris: L'lnstitut d'Ethnologie.
1987 [1615] Nueva coronica y buen gobierno. Edited by John V. Murra, Rolena Adomo,
and Jorge L. Urioste. Madrid: Historia 16.
HARDMAN-DE-BAUTISTA, MARTHA JAMES
1982
"The mutual influence of Spanish and Andean languages." Word: ]ournal of the
Linguistic Circle of New York 33 (1-2): 143-57.
JARA, VICTORIA DE LA
1970
"La solucidn del problema de la escritura peruana." Arqueologia y sociedad 1:2735.
JULIEN, CATHERINE
1988
2000
"How Inca Decimal Admirustration Worked." Ethnohistory 35 (3): 357-79.

Reading Inca History. Iowa City: Uruversity of Iowa Press.
LEONARD, IRVING
1942
"Best Sellers of the Lima Book Trade, 1583." Spanish American Historical Review
22: 5-33.
LOCKE, L. LELAND
1923
The Ancient Quipu or Peruvian Knot Record. New York: American Museum of
Natural History.
LOCKHART, JAMES
1985
"Some Nahua Concepts in Post-Conquest Guise." History of European Ideas 6 (4):

465-82.
L6PEZ GRIGERA, LUISA
2001
"Introduccidn al estudio ret6rico de La Primer Coronica atribuida a Guaman

Poma." In Guaman Poma y Bias Valera: Tradicion Andina e Historia Colonial, edited
by Francesca Cantu, 273-91. Rome: Instituto Italo-Latinoamericano.
MACKEY, CAROL
1990
"Comparaci6n entre quipu inca y quipu modemps." In Quipu yyupana: coleccion

de escritos, edited by Carol J. Mackey et al., 135-55. Lima: Consejo Nacional de
Ciencia y Tecnologia.
MAZZOm, ANTONIO
1996
Coros Mestizos del Inca Garcilaso. Lima: Fondo de Cultura Economica.
I45
MCLUHAN, MARSHALL
1962
The Gutenberg Galaxy: The Making of Typographic Man. Toronto: University of

Toronto Press.
MENDIZABAL LOSACK, EMILIO
1963
"Las dos versiones de Murua." Revista del Museo Nacional 32:153-85.
MIGNOLO, WALTER D.
1994
1995
"Afterward: Writing and Recorded Knowledge in Colorual and Post-Colonial

Situations." In Writing without Words: Alternative Literacies in Mesoamerica and
the Andes, edited by Elizabeth Hill Boone and Walter D. Mignolo, 292-313.
Durham: Duke University Press.
The Darker Side of the Renaissance: Literacy, Territoriality, and Colonization. Ann
Arbor: University of Michigan Press.

MOLINA, CR1ST6BAL DE
1988 [1576] Pabulas y mitos de los Incas. Madrid: Historia 16.

MROZ, MARCIN
1984
1989
"Una interpretacion numerica de la cr6nica de Guaman Poma de Ayala."

Anthropologica 2 (2): 67-103.
"Paracronologia dinastica de los Incas segun Guaman Poma de Ayala." In Time
and Calendars in the Inca Empire, edited by Mariusz S. Ziolkowski and Robert M.
Sadowski, 17-34. Oxford: BAR International Series, 1989.
MURRA, JOHN
1968
1981
"An Aymara Kingdom in 1567." Ethnohistory 15 (2): 115-51.

"Las etnocategorias de un khipu estatal." In La tecnologia en el mundo andino,
edited by Heather Lechtman and Ana Maria Soldi, 433-42. Mexico, D.F.:
Universidad Nacional Aut6noma de Mexico.
MUROA, MARTIN DE
1986 [1611] Historia general del Peru. Edited by Manuel Ballesteros. Madrid: Historia 16.
NILES, JOHN D.
1999
Homo Narrans: The Poetics and Anthropology of Oral Literature. Philadelphia: University of Pennsylvarua Press.
O'GORMAN, EDMUNDO
1995 [1958] La invencidn de America. Mexico: Fondo de Cultura Econbmica.

OLSON, DAVID R., AND NANCY TORRANCE
1996
Modes of Thought: Explorations in Culture and Cognition. Cambridge: Cambridge

University Press.
ONG, WALTER
1982
Orality and Literacy: The Technologizing ofthe Word. New York: Routledge.
OSSIO, JUAN
1970
1973
1998
"The Idea of History in Felipe Guaman Poma de Ayala." B. Litt. thesis, Oxford
Uruversity.
"Guaman Poma. Nueva cor6ruca o carta el rey: Un intento de aproximacion a
las categorias del pertsamiento andino." In Ideologia mesianica del mundo andino,
edited by Juan M. Ossio, 153-213. Lima: Ignacio Prado Pastor.
"El original del manuscrito de Loyola de Fray Martin de Murua." Colonial Latin
American Review 7 (2): 271-78.
PARSSINEN, MARTTI
1992
The Inca State and Its Political Organization. Helsinki: Societas Historica Finlandiae.
P S R E Z CANT6, MARlA PILAR
1996
El Bueno Gobiemo de don Pelipe Guaman Poma de Ayala. Quito: Abya-yala.
PfiREZ DE GUZMAN, F E R N A N
1965 [1450] Generaciones y semblanzas. Edited by Robert Bryan Tate. London: Tamesis.
PLAS, SOFIE
1996
"Une source europeenne de la Nueva coronica y buen gobiemo de Guaman Poma."

Journal de la Societe des Americanistes 82:97-116.
PORRAS BARRENECHHA, RAtJL
1986
"Cronicas olvidadas sobre el incario." In Los cronistas del Peru (1528-1650), edited by Franklin Pease G.Y., 747-51. Lima: Banco de CrMito del Peni.
PRADA RAMfREZ, FERNANDO
1995
"El khipu incaico: de la matem^tica a la historia." Yachay 12 (21): 11-37.
146
PRATT, MARY LOUISE
1992
Imperial Eyes: Travel Writing and Transculturation. New York: Routledge.
PULGAR, HERNANDO DEL
1971 [1486] Claros varones de Castilla. Edited by Robert Brian Tate. Oxford: Clarendon
Press.
QUISPE AGNOU, ROCfO
2002
"Escritura alfab^tica y literalidades amerindias: fundamentos para una

historiografia colonial andina." Revista andina 34: 237-52.
RADICATI DI PRIMEGLIO, CARLOS
1965
1979
"La seriaci6n como posible clave para descifrar los quipus extranumerales."
Documenta: Revista de la Sociedad Peruana de Historia 4:112-215.
El sistema contable de los Incas: yupana y quipu. Lima: Libreria Studium.
RODRfGUEZ DE ALMELA, DIEGO
1994
Valerio de las estorias escoldsticas e de Espana. Edited by Juan Torres Fontes. Murcia:
Real Academia Alfonso X El Sabio, [1487].
ROMERO, ](Xt LUIS
1945
"Sobre la biografia espafiola del siglo XV y los ideales de la vida." In Sobre la

biografia y la historia, 47-87. Buenos Aires: Editorial Sudamericana.
ROSTWOROWSKI, MARiA
1990
"La visita de Urcos de 1562: Un kipu pueblerino." Historia y cultura 20:295-317.
ROWE, JOHN H.
1958
1985
"The Ages Grades of the Inca Census." Miscellanea Paul Rivet Octogenario Dicata,
vol. 2,499-522. Mexico, D.F.: Universidad Nacional Autonoma de Mexico.
"Probanze de los Inca nietos de conquistadores." Historica 9 (2): 193-245.
SANCHEZ ALONSO, BENITO
1941
Historia de la historiografia espafiola. Madrid: Consejo Superior de Investigaciones

Cientificas.
TATE, ROBERT BRIAN
1965
Introduction to Generaciones y semblanzas, by Femdn P^rez de Guzman, vii-xxvii.

London: Tamesis.
TODOROV, TZVETAN
1977
1981
The Poetics of Prose. Ithaca: Cornell University Press.

Introduction to Poetics. Minneapolis: Uruversity of Minnesota Press.
TURNER, MARK
1996
The Literary Mind: The Origins of Thought and Language. New York: Oxford University Press.
URTEAGA, HORACIO
1921
Introduction to Discurso sobre la descendencia y gobiemo de los Incas, by Collapina

and Supno and other Khipukamayuqs. In Informaciones sobre el antiguo Peru, edited by Horacio Urteaga. Lima: San Marti y Cia.
URTON, GARY
1990
1994
1997
1998
2001
2002
The History ofa Myth: Pacariqtambo and the Origen ofthe Inkas. Austin: Uruversity
of Texas Press.
A "New Twist in an Old Yard: Variation in Knot Directionality in the Inka
Khipus." Baessler-Archiv Neue Polge 42: 271-305.
The Social Life of Numbers: A Quechua Ontology of Numbers and Philosophy of Arithmetic. Austin: The University of Texas Press.
"From Knots to Narratives: Reconstructing the Art of Historical Record Keeping in the Andes from Sparush Transcriptions of IrJca Khipus." Ethnohistory 45
(3): 409-38.
"A Calendrical and Demographic Tomb Text From Northern Peru." Latin American Antiquity 12 (2): 127-47.
"Recording Signs in Narrative-Accounting Khipu." In Narrative Threads: Accounting and Recounting on Andean Khipu, edited by Jeffrey Quilter and Gary Urton,
171-96. Austin: University of Texas Press.
VAZQUEZ ESFINOSA, ANTONIO
1992 [1630] Compenio y descripcion de las Indias Occidentales. Edited by Balbino Velasco
Baydn. Madrid: Historia 16.
I47
VEGA, JUAN JOS
1974
Prologue to Relacion de la descendencia, gobierno y conquista de los Incas. Lima:
Ediciones de la Biblioteca Universitaria.

WACHTEL, NATHAN
1971
1973
La vision des vaincus. Paris: Gallimard.

"Pensamiento salvaje y aculturacion: el espacio y el tiempo en Felipe Guaman
Poma de Ayala y el Inca Garcilaso de la Vega." In Sociedad e Ideologia: Ensayos de
historia y antropologia andinas. Lima: Instituto de Estudios Peruanos.

Brokaw (2003) The Poetics of Khipu Historiography

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Brokaw (2003) The Poetics of Khipu Historiography

Uploaded by

Copyright:

Available Formats

THE POETICS OF KHIPU HISTORIOGRAPHY:

Felipe Guaman Poma de Ayala's Nueva coronica

Latin American Research Review

Latin American Research Review

characterized the original indigenous text. Even though these writers

Incas (1542/1608) or produced directly by indigenous writers who claim

Latin American Research Review

TEXTS AND CONTEXTS

The Nueva coronica was written by Felipe Guaman Poma de Ayala,*

Poma [1615] 1987,8[8]).

explicit in terms of its relationship to khipu sources, but its provenience

Latin American Research Review

Latin American Research Review

or lesser extent structures the accoimt (Julien 2000). The identification of

Prior to the conquest of the Americas, Spanish historiographers were

"Historiografia medieval: constantes evolutivas de un genero," Fernando

Latin American Research Review

Table 1 Discursive Paradigm ofthe European Biographical Genre

Undeniably European discursive genres influenced Guaman Poma's

Latin American Research Review

Any given genre or discursive formation is tied to an ideology. The

introductory material not present in the other biographies. Perhaps not

bivid mds de veinte e siete aiios. (P^rez de Guzman [1450] 1965,4-5)

Latin American Research Review

Table 2 lists the essential components of Guaman Poma's biographies

Latin American Research Review

Table 2 Narrative Outline of Guaman Poma's Biographies

Distinctive dress of the Inca

No real discussion of temperament takes place, and the moralizing tone

afios. (CoUapina and Supno [1542/1608] 1921,13)

Latin American Research Review

Table 3 Format of the Discurso

Names ruler as successor to previous emperor

Although a European reading of the Discurso may identify it with the

1. Dress of the Inca

phies link the transitional phrase to the following biography: "Lluqui

might have been the work of the transcriber/translator. Horacio

Latin American Research Review

Latin American Research Review

In Julien's comprehensive survey of colonial chronicles, she argues

Latin American Research Review

Poma [1615] 1987, 111 [111]).

Latin American Research Review

Latin American Research Review

semiosis as it expanded to accommodate the needs of the Inca social

In the context of a cultural contact zone such as the colonial Andes,

Latin American Research Review

Although this type of comparative analysis is very productive, it is

La Nueva coronica y buen gobiemo: lObra de Guaman Poma 0 de jesuitas?

ALTIERl, ANDRfe RADAMES

"Sobre un kipu peruano." Revista de Antropologia de la Universidad Nacional de

ARELLANO HOFFMAN, CARMEN

"Quipu y tocapu: Sistemas de comurdcacion Inca." In Los Incas: Arte y simbolos,

"The Logical-Numerical System of the Inca Quipus." Annals of the History of

ASCHER, MARCIA, AND ROBERT ASCHER

"Code of Ancient Peruvian Knotted Cords (Quipus)." Nature 222: 529-33.

"Inka Writing." In Narrative Threads: Accounting and Recounting in Andean Khipu,

BOONE, ELIZABETH HILL

"Introduction: Writing and Recording Knowledge." In Writing without Words:

1985 [1553] E; senorio de los Incas. Madrid: Historia 16.

COLLAPINA, AND SUPNO, ET AL.

1921 [1542/1608] "Discurso sobre la descendencia y gobierno de los Incas." In

^^*'" American Research Review